Mechanical site preparation for forest restoration

Magnus Lof • Daniel C. Dey • Rafael M. Navarro • Douglass F. Jacobs

Received: 21 December 2011 / Accepted: 9 April 2012 / Published online: 22 April 2012� Springer Science+Business Media B.V. 2012

Abstract Forest restoration projects have become increasingly common around the

world and planting trees is almost always a key component. Low seedling survival and

growth may result in restoration failures and various mechanical site preparation tech-

niques for treatment of soils and vegetation are important tools used to help counteract this.

In this article, we synthesize the current state-of-knowledge concerning mechanical site

preparation for improved tree establishment when carried out in different forest restoration

situations, point out critical research gaps and provide some recommendations for future

directions. Mechanical site preparation often results in improved seedling survival and

growth. However, if not intensive methods with much soil disturbance are used, it is a

rather ineffective tool for controlling competing vegetation. Methods such as scarification,

mounding and subsoiling also lead to multiple interactions among soil physical and

chemical properties that affect plant survival and growth, and it may be difficult to

determine the actual cause–effect relationship of any positive seedling responses. Most

research to date on mechanical site preparation and plantation performance has been

conducted using a few conifer tree species. Seedling responses differ among tree species

and alternative species are often used during restoration compared to production forestry

M. Lof (&)Southern Swedish Forest Research Center, Swedish University of Agricultural Sciences,P.O. Box 49, 230 53 Alnarp, Swedene-mail: magnus.lof@slu.se

D. C. DeyU.S. Forest Service, Northern Research Station, P.O. 202, Anheuser-Busch Natural Resources Bldg.,Columbia, MO 65211, USA

R. M. NavarroSchool of Agricultural and Forestry Engineering, University of Cordoba, Edf. Leonardo da Vinci,Campus de Rabanales, 14071 Cordoba, Spain

D. F. JacobsDepartment of Forestry and Natural Resources, Hardwood Tree Improvement and RegenerationCenter, Purdue University, West Lafayette, IN 47907-2061, USA

123

New Forests (2012) 43:825–848DOI 10.1007/s11056-012-9332-x

indicating a need for additional research for improved understanding. Several management

objectives such as soil protection and increased biodiversity are many times relevant

during forest restoration, and mechanical site preparation methods should be implemented

carefully because they can have large impacts on the environment.

Keywords Afforestation � Ecosystem management � Rehabilitation � Regeneration �Sustainability

Introduction

The tree regeneration phase offers the best opportunity to change tree species and forest

ecosystem structure, and this phase is an important initial step in restoration of forests.

Without measures to improve soil conditions, control competing vegetation and reduce

animal damage, artificial or natural forest regeneration often results in unacceptably low

seedling performance. Low seedling performance, i.e. survival and growth, may result in

substantial economic losses and various site preparation techniques are important tools to

counteract this (e.g. Nyland 1996). This is usually achieved by using herbicides, mechanical

site preparation, prescribed burning or mulching, which can be used singly or in combi-

nation (Prevost 1992; Sutton 1993; Iverson et al. 2008; Willoughby et al. 2009). Prescribed

burning is limited seasonally by weather and fuel conditions, usually requires multiple

applications, and may exacerbate competition problems when undesired or invasive veg-

etation is adapted to fire (Iverson et al. 2008; Rebbeck 2012). Mulching is rather expensive

and is more common in urban plantations and in horticulture than in forest restoration

projects. In general, forest managers prefer using herbicides to control competing vegetation

because of their effectiveness and relatively low cost compared to available alternatives

(Willoughby et al. 2009). However, there are substantial differences between countries and

continents with regard to preferred site preparation methods (Ammer et al. 2011). In many

European countries for example, public acceptance of chemical herbicides is low and

certification systems generally discourage their use, prompting managers to examine

alternative management practices such as mechanical site preparation (MSP) (Thiffault and

Roy 2011). In addition to vegetation control, MSP may also be used as an important tool for

improving physical soil conditions that limit tree regeneration or restoring site hydrology.

MSP is a broad category of site preparation methods typically involving use of large

heavy machinery with certain attached implements to prepare an area and its soil for tree

regeneration. Previously, MSP has mostly been used for increased productivity in tradi-

tional plantation forestry (Orlander et al. 1990). However, during the past decades there has

been an acceleration of forest restoration activities around the globe (e.g. Stanturf and

Madsen 2005; Perrow and Davy 2008), in which MSP methods are widely used. MSP is

used for land reclamation following use for agricultural or surface mining operations, and

for forest rehabilitation (e.g. Moffat and Bending 2000; Van Lerberghe and Balleux 2001;

Querejeta et al. 2001; Parotta and Knowles 2001; Lof et al. 2006). See Stanturf (2005) for

various concepts connected to forest restoration. While effects of MSP on regeneration in

traditional plantation forestry is well documented and summarized (e.g. Orlander et al.

1990; Prevost 1992; Sutton 1993), no thorough synthesis of this topic has been conducted

specific to use of MSP in forest restoration. As MSP can dramatically influence forest

restoration outcomes, the aim of this review is to summarize and synthesize our current

knowledge about effects on seedling performance in relation to MSP and to evaluate

economic efficiency of MSP and its environmental impacts during forest restoration.

826 New Forests (2012) 43:825–848

123

This review focuses on MSP for treatment of soils and vegetation by the three main

types; scarification, mounding, and sub-soiling/ripping that are commonly used today.

Although many times the first step in restoration scenarios, MSP types for only above

ground purposes such as mowing, drum chopping, blading and piling, which are mainly

used to clear a site from vegetation and slash, are not included in this review. In addition,

intensive MSP methods such as deep plowing and terracing, which due to environmental

reasons are seldom used nowadays, are touched upon only briefly.

Mechanical site preparation methods

There are several purposes for using MSP methods in advance of regenerating a site.

Depending on the equipment used, MSP can include cultivating the soil layers: clearing

unwanted vegetation and removing logging slash to facilitate planting or direct seeding

(Nyland 1996). Depending on the site conditions, MSP may involve manipulating the soil

conditions to improve microsites for seedlings by increasing or decreasing soil moisture as

needed, increasing solar radiation, modifying soil temperature, increasing soil nutrient

availability, reducing soil compaction and controlling competing vegetation (Prevost

1992). In addition, MSP may be used to reduce potential damaging effects from insects

such as the pine weevil (Hylobius abietis) or from small mammals and birds feeding on

seeds (Lof 2000; Birkedal et al. 2010).

Because large machinery most often is needed for MSP, it is most efficient in treating

large areas and is therefore best suited for even-aged regeneration systems using artificial

regeneration. For example, restoration of modern surface mining sites can be on an

enormous scale including several square kilometers (Bradshaw and Huttl 2001). For some

tree species such as European beech (Fagus sylvatica), however, MSP is also used in

uniform shelterwood systems to enhance natural regeneration by improving seedbed

characteristics (Madsen 1995). MSP methods are many times used in combination to

achieve multiple purposes. For example, sites prepared for conversion of Norway spruce

(Picea abies) to broadleaves in southern Sweden are normally treated by piling of slash to

ready the site for planting, which is then followed by disk trenching to prepare the soil (Lof

et al. 2004). In Spain, terracing of hill slopes may be followed by subsoiling to improve

water infiltration into the soil and plant root development (Querejeta et al. 2001). In

afforestation of bottomland pastures or old fields in the US, mowing of vegetation may be

followed by soil ripping, disking or mounding (Allen et al. 2004; Kabrick et al. 2005).

MSP for treatment of soils and vegetation can be carried out in three different ways or

intensities; (1) continuous as connected tracks or areas, (2) intermittent with patches,

mounds or subsoiling done at regular distances, or (3) directed where patches or mounds

are placed at certain suitable areas. The choice of method may depend on soil type,

topography and the degree of disturbance needed for regeneration or other restoration or

conservation concern. For example, directed MSP may be selected in steep terrain or where

rocky conditions prevent continuous or intermittent methods and where there is risk of soil

erosion or other negative environmental impacts.

Site preparation for forest regeneration has essentially been practiced since humans

began managing forests. Wild boars (Sus scrofa), for example, were domesticated several

thousand years ago and the use of pannage (pasturing in a forest) to feed pigs was recorded

from the antiquity in Europe and still has not totally disappeared. This practice facilitates

beech and oak (Quercus spp.) natural regeneration (Peters 1997). In addition, prescribed

burning has long been used for improved natural and artificial regeneration (Nyland 1996).

New Forests (2012) 43:825–848 827

123

Several of the MSP methods (see below) commonly used during the last 50 years were

already described in detail 150 years ago (e.g. Orlander et al. 1990; Sutton 1993 and

references therein), although at that time manual or horse power were used instead of

heavy machines. Following the introduction of the clear-cutting system in northern

Scandinavia around 1950, various methods of MSP were developed and gradually intro-

duced in practical forestry (Soderstrom 1976; Ritari and Lahde 1978). During the same

time period, MSP methods were introduced for restoration of former mine sites and of

Mediterranean dry sites (e.g. Moffat and Bending 2000; Barbera et al. 2005), and with the

ready availability of gas and diesel machinery in the 1940s, MSP became integrated into

forestry practices especially in even-aged management systems and intensive plantation

culture. The three main types of MSP for treatment of soils and vegetation (i.e. scarifi-

cation, mounding, and subsoiling/ripping) that are commonly used today are described

below (Fig. 1). Depending on machinery and attached equipment they can all be carried

out using different intensity as described above.

Scarification of the soil to remove vegetation and the upper organic layers and uncover

bare soil can be achieved through several different techniques. Patch scarification (inter-

mittent) and disc trenching (continuous) are among the most commonly used in boreal

forests (Luoranen and Rikala 2012) (Fig. 1). Chaining, i.e. dragging heavy chains to

remove vegetation and organic layers has approximately the same effect on the soil as disc

trenching. Scalping is another word for patch scarification, although it is more related to

the removal of grass sod to uncover bare soil during afforestation. Blading using mounted

blades to uproot trees and scrubs may also be listed under scarification methods if the soil

is also treated; it is then a continuous method that uncovers relatively large areas of bare

soil. The purpose of scarification MSP techniques is to create desirable planting spots in

mineral soil or in mixed mineral-organic soil. In addition, increased soil temperature and

moisture resulting from this treatment provides advantages for seedling growth (Prevost

1992). Planting spots are normally below the original ground level, therefore temporary

standing water may occur (Prevost 1992). Thus, different planting spots may be chosen

depending on, for example, the moisture regime at the site (Boateng et al. 2011). Other-

wise, scarification does not significantly influence soil structure.

In contrast to scarification techniques, mounding creates elevated planting spots and

influences soil structure (Fig. 1) (Sutton 1993). Spot-wise mounding (intermittent) removes

vegetation and the upper organic levels and results in inverted or mixed soil on top of

either the organic layer or bare soil. Bedding is a continuous form of mounding to create

beds with an elevated center. Plowing is another, more intensive, form of MSP methods

that may be listed under continuous mounding because it also creates elevated planting

A B C

Organic

Mineral

Fig. 1 Schematic description of three types of mechanical site preparation and their main effect on soilstructure: scarification (A), mounding (B) and subsoiling (C). Dark grey area below mineral soil representhardpan in C. Normal planting spots are indicated with a seedling. For A and B some authors recommendplanting at the trench-berm interface (e.g. Boateng et al. 2011)

828 New Forests (2012) 43:825–848

123

spots on ridges. The purposes of mounding MSP techniques are to create elevated planting

spots free from water logging and with little vegetative competition. Also, increased soil

temperature and nutrient availability, reduced soil bulk density, and good soil aeration are

mentioned as advantages for root and seedling growth (Ritari and Lahde 1978; Londo and

Mroz 2001; Kabrick et al. 2005). Mounds of different sizes may be created depending on

the moisture regime at the site. Inverted site preparation, or hole-planting, where the

mound is put back in the dug-out hole, are also sometimes listed under mounding tech-

niques although the planting spot normally is not significantly elevated (Sutton 1993)

(Fig. 2). Sometimes the latter technique is also called spot-tillage, or complete cultivation

if the treated area is larger.

Subsoiling or ripping is a MSP method used for dry soils, reclaimed mined sites or for

soils that have a compacted surface layer, but more often below the soil surface, that

restricts root growth and plant development (e.g. Moffat and Bending 2000; Barbera et al.

2005; Palacios et al. 2009) (Fig. 1). Compacted layers may develop after long-term grazing

regimes, working of soils during mining operations or as hardpans following agricultural

land use. Without subsoiling, plantations may be stagnate after a few years despite positive

early growth and establishment (Van Lerberghe and Balleux 2001). Subsoiling also is used

to increase soil aeration, soil water infiltration and drainage, and to reduce soil bulk

density. Subsoiling fractures soil structure without mixing of soil horizons and it is many

times the first stage in a two-step site preparation process that also involves use of her-

bicides or other MSP methods to control vegetation and create suitable microsites for

regeneration (Stanturf et al. 2004; Gwaze et al. 2007) (Fig. 3).

Fig. 2 Inverting site preparation by excavator (directed MSP) in southern Sweden to prepare planting spotsfor beech seedlings during conversion of Norway spruce to mixed broadleaved forests dominated by beech.Photo by Jorg Brunet

New Forests (2012) 43:825–848 829

123

Casual factors behind seedling performance

Mechanical site preparation (MSP) types such as scarification, mounding and subsoiling

influences both soil conditions and the intensity of competing vegetation, thus, the specific

cause of any positive effects on seedling development is often not well elucidated (Mar-

golis and Brand 1990). In addition, confounding effects such as altered herbivory by

insects or browsing and predation from small mammals makes it even more difficult to

correctly interpret survival and growth responses (Lof 2000; Birkedal et al. 2010). These

responses may also vary with climatic conditions, site types and tree species (Prevost

1992). Thus, the choice of method in various forest restoration situations may be prob-

lematic; and depend on existing biotic and abiotic conditions and expected problems

following any treatment sequence.

In general, earlier research (e.g. Barring 1965; Soderstrom 1976; Soderstrom et al.

1978) on MSP and plant responses showed that performance of planted seedlings improves

with increasing intensity of MSP. However, neither interfering vegetation nor insect

damage was fully controlled in these early pioneering experiments and, thus, the conclu-

sions regarding casual factors behind plant responses were rather speculative. In a classical

study conducted during the late 1980s, Munson et al. (1993) showed the full influence from

interfering vegetation on plant responses in connection with MSP. When vegetation was

continuously controlled, i.e. every year, there was almost no additional positive growth

effect in seedlings from scarification (Fig. 4). In the same experiment, there was also no

additional effect of continuous fertilization. Therefore, and at many sites, any positive

effects from increased soil temperature and mineralization rates during pure scarification

seem minor. At most sites, the main influence on planted seedling performance following

Fig. 3 Subsoiling (ripping) operation (continuous MSP) on a former agricultural field in the LowerMississippi Alluvial Valley, southern USA. The rig is spraying a band of herbicide over the ripped row.Photo by Ray Souter

830 New Forests (2012) 43:825–848

123

scarification is decreased interference from vegetation, an effect that may be important

during the critical seedling establishment period, but which is rather short-lived as exposed

soil is rapidly re-colonized with vegetation. However, the above mentioned study was not

carried out in the boreal forest where sometimes the humus layer may be 20–30 cm thick.

For example, Thiffault et al. (2010) reported similar growth responses in planted seedlings

from scarification and herbicide treatments 15 years following planting at a boreal site in

eastern Canada.

Scarification may also sometimes improve conditions for germination of unwanted

herbaceous and woody vegetation and, thus, aggravate the negative interference from

surrounding vegetation on seedlings (Prevost 1992). Horsley (1991) found that mechanical

application of herbicides by tracked vehicles to control competing vegetation in the

understory of Allegheny hardwood forests actually caused high levels of competition after

the shelterwood regeneration harvest was done. The metal cleats on the vehicle tracks

segmented fern rhizomes that rendered ineffective the glyphosate treatment. Ferns pro-

liferated after harvesting and dominated the ground flora to the detriment of the desired

hardwood regeneration.

Although not the most efficient vegetation control method, scarification MSP may have

other positive effects on plant performance. In Europe and parts of Asia, the pine weevil is

a well-known destructive insect in reforestations, especially on conifer sites where, without

protective measures, it may cause unacceptably high mortality of conifer seedlings fol-

lowing planting (Lof 2000). MSP by scarification provides some protection as long as bare

soil surrounds the seedlings (Petersson et al. 2005). MSP by mounding provides better

protection against the weevil by causing more soil disturbance, although continuous

insecticide treatment provides the best protection (Orlander and Nilsson 1999). This effect

of pine weevil damage on tree seedlings performance is tree species and site dependent,

however, and they prefer to feed on conifer tree species much more so than on broadleaved

tree species (e.g. Lof et al. 2004). In addition, outside its distribution area, on former

agricultural land or on former broadleaved forest sites, the pine weevil is not a problem for

regeneration and, thus, not relevant at all forest restoration sites.

Fig. 4 The influence ofcombinations of scarification (S),repeated annual fertilization(F) and repeated annual herbicidetreatment (H) and without anytreatment (C) on the basal areaincrement on Pinus strobusseedlings during the 4 yearsfollowing planting. Redrawnfrom Munson et al. (1993) wherethe experiment was conductednear Ottawa, southern Canada

New Forests (2012) 43:825–848 831

123

Enhanced seed survival and germination during natural regeneration or direct seeding of

a number of tree species is another example of the positive effects of scarification and

mounding (Karlsson 2001; Madsen 1995; Lof and Birkedal 2009). Contact with bare soil

facilitates water uptake by seeds and small seedlings, which otherwise may suffer desic-

cation in the organic layer, and therefore enhances both seed and seedling survival (Prevost

1992). As seed production varies substantially between tree species and years, and because

patches of bare soil are rapidly re-colonized with vegetation, the effect of MSP on natural

regeneration depends on when the operation is carried out and in which ecosystem

(Karlsson 2001). A unique application of scarification in conservation of oak forests in the

US is to increase the number of acorns that get incorporated into the soil and organic layers

of forest understories, and thereby increase the number of established seedlings by pro-

tecting the acorns from desiccation, and perhaps reducing their susceptibility to predation

by small mammals. After a good acorn crop was on the ground, Lhotka and Zaczek (2003)

used a small brush rake mounted on a crawler dozer to substantially reduce the density of

the sugar maple midstory while incorporating acorns in the upper soil layer. Similarly,

Rathfon et al. (2008) used a small tractor-pulled disk to scarify the upper soil layer in the

understory of oak forests to incorporate acorns that were on the ground. They followed this

with a combination of herbicide treatments to reduce the midstory density. In both of these

studies, scarification significantly increased the density of new oak seedlings and midstory

removal improved oak survival. Finally, since MSP may improve conditions for seed

germination of a number of herbaceous and woody species, it may also increase biodi-

versity (see below under impacts on environment).

The relative importance among growth factors differs between regions and this influ-

ences which MSP method is preferred. The Boreal forest is characterized by low pro-

ductivity, primarily resulting from a short growing season, low temperatures and poor soil

nutrient availability (Tamm 1991). In contrast, water becomes more limiting to plant

growth in the Temperate, Mediterranean and Tropical zones, where growing seasons are

longer and higher temperatures promote greater mineralization rates in more productive

habitats (Kramer and Boyer 1995; Blondel and Aronson 1999). Thus, mounding may

create elevated planting spots with increased soil temperatures and improve soil drainage

and moisture conditions for seedlings at northern boreal sites (Sutton 1993), whereas on

Mediterranean dry sites, elevating planting spots created by mounding and fertilization

may be negative for seedling performance (Varelides and Kritikos 1995). In the Medi-

terranean region, subsoiling may have been a better choice of MSP method to increase the

amount of available soil water. Changing climatic conditions between years and variation

between site types including amounts of vegetation may also affect the outcome of plant

responses following MSP (Nilsson and Orlander 1995; Gemmel et al. 1996).

In contrast to scarification, and in experiments at relatively wet or cold sites where

vegetation has been controlled, mounding itself may have a more significant positive effect

on seedling growth (Sutton 1993; Lof et al. 2006). In general, this growth response has

been correlated to increases in soil temperatures, available soil nutrients, soil aeration, and

nutrient seedling uptake, yet decreased soil moisture and soil density (Knapp et al. 2006;

Lof and Birkedal 2009). The specific cause of positive effects in seedlings, however, is

likely difficult to determine. Alternatively, it is the combination of changes that causes the

effect. Similar to mounding, subsoiling may have a positive effect on plant performance

through increased rooting depth and water uptake at drier sites (e.g. Querejeta et al. 2001;

Palacios et al. 2009), and improved drainage at compacted or wet sites (Van Lerberghe and

Balleux 2001). In contrast to scarification and mounding, subsoiling does not reduce

competition from herbaceous vegetation (Querejeta et al. 2001).

832 New Forests (2012) 43:825–848

123

Most research investigating MSP and plant performance has been conducted using a

few conifer tree species with the goal of increased productivity in traditional plantation

forestry or during restoration where the conifer species is looked upon as a relatively easily

established nurse tree species to improve soils and the environment for other tree species

(e.g. Prevost 1992; Sutton 1993; Pausas et al. 2004). The outcome in terms of seedling

performance following MSP is very much dependent on tree species, site and situation.

Furthermore, according to our knowledge there are no replicated experimental investiga-

tions that evaluate the effects of MSP throughout several vegetation zones. Our under-

standing of MSP and plant performance is therefore limited to a few tree species and

knowledge is lacking regarding seedling performance following various MSP methods in

relation to climate. If the goal of forest management is to implement MSP on a broader

scale for forest restoration, research in these areas becomes more important. To better

understand the causes for any positive seedling responses to MSP, this research should

include a wide spectrum of sites and tree species.

Forest restoration and seedling performance

Afforestation of former farmlands

Human population increase and degradation of forests over large areas is normally fol-

lowed by cultivation of agricultural crops on marginal lands, not well suited for sustainable

crop production (Weber 2005). Afforestation by artificial regeneration or by natural

invasion of trees often occurs first on those marginal sites as a consequence of land use

change, shifts in agriculture practices and depopulation of rural areas. Thus, afforestation

of former farmland is carried out on a variety of soil types, including dry and wet sites in

uplands and floodplains. Although land is considered marginal from an agricultural per-

spective, sites may still be productive and competition between herbaceous vegetation and

newly established seedlings is many times a main problem for successful afforestation

(Van Lerberghe and Balleux 2001; Navarro Cerrillo et al. 2005). As cropfields revegetate

damage to trees may occur from small mammals, especially voles, who find suitable

habitats among the weeds (Barring 1967). Another problem for successful afforestation

may be plow-pans or deeper layers of clay that restricts root growth (Stanturf et al. 2004).

With water logged soils it may be necessary to construct a system with open ditches to

remove excess water before planting (Van Lerberghe and Balleux 2001). Deep plowing

down to 60–70 cm may also be used to remove excess water through furrows. Elevated

planting spots are then available on top of the ridges formed from construction of the

furrows. This may be combined with subsoiling to further increase soil drainage, and with

a system of ditches. In the latter case, the ditches must be dug after plowing, so as not to

hamper the plowing, and to ensure that the furrows run directly into the ditches. In cases

where soil texture itself causes low permeability, i.e. soils with fine texture and clay

content, it may be necessary to further reshape the soil surface. Mounding, sometimes

called bedding, is typically performed utilizing plows where soil is turned inward from two

directions to create elevated planting spots. There are several studies showing improved

seedling performance, i.e. survival and growth, with use of this method on former farmland

having excess water (e.g. Patterson and Adams 2003; Stanturf et al. 2004). However,

bedding does not always result in improved seedling performance where coarse textured

soils are naturally well-drained (Kabrick et al. 2005).

New Forests (2012) 43:825–848 833

123

If plantations are established on farmland immediately removed from agricultural

production and on moderately fertile soils, MSP is often not practiced because herbaceous

vegetation has not yet colonized the sites (Gardiner et al. 2002). However, subsoiling may

be needed to break plow-pans where they have developed (Stanturf et al. 2004). For

example, subsoiling may improve soil moisture availability, root development and soil

exploitation for planted seedlings (Miller 1993; Bocio et al. 2004; Dey et al. 2008). In

addition, seedlings may tolerate competition from herbaceous vegetation better after

subsoiling and significant increases in seedling survival and growth have been found for

several tree species following implementation of this MSP method on former farmland

(Berry 1986; Russell et al. 1997; Self et al. 2010). However, subsoiling on former farmland

does not always improve seedling performance and the outcome may depend on soil type

and climate (du Toit et al. 2010). Where fields have been removed from agricultural use for

several years prior to regeneration, MSP methods such as disking or deep plowing are

practiced to reduce competition from herbaceous vegetation and remove small mammal

habitat (Gardiner et al. 2002; Orlander et al. 2002).

Post planting MSP using harrowing of topsoil between planting lines to reduce com-

petition is sometimes used, but needs to be carried out every year to improve seedling

performance (Cogliastro et al. 1997). For example, Seifert et al. (1985) observed that

disking for two consecutive years, the year before and after planting, or using a combi-

nation of disking for 2 years and raised beds improved height and diameter increment of

planted 1–0 swamp chestnut oak (Quercus michauxii Nutt.) on poorly drained flat pastures

in southeastern Indiana, US. In contrast, where Groninger et al. (2004) used only a single

disking to prepare a bottomland former cropfield for the planting of 1–0 green ash

(Fraxinus pennsylvanica Marsh.), they found that it did not affect seedling height or

diameter growth after 3 years in southern Illinois, US. Others have reported that MSP by

disking upland pastures, old fields and deforested areas once did not improve survival or

growth of planted hardwood bareroot seedlings, and infact promoted the development of

competing vegetation (Horsley 1985; Jacobs et al. 2004). Multiple cultivations with a disk

in a single season did generally improve initial height and diameter growth of 1–0 planted

stock compared to no MSP, but the significance of the difference depended on the tree

species in an afforestation study in southeastern Indiana, US (Seifert and Woeste 2002).

MSP methods that remove top soil, i.e. deep plowing, and create seedbeds with mineral

soil have also been shown to improve natural regeneration of birch (Betula spp.) on former

farmlands in Sweden, whereas MSP treatments that preserved top soil enhanced germi-

nation of herbaceous species (Karlsson 1996).

Afforestation using deep plowing to 60–70 cm soil depth on poor sandy soils was

common practice in western Denmark (Neckelmann 1976) and Mediterranean countries

(Navarro-Garnica 1975). Due to burial of topsoil containing seeds from weeds, burial of

organic matter with high water holding capacity and breakage of any plow pans, survival

and growth of seedlings are normally improved (Madsen et al. 2005). However, this MSP

method is now seldom used due to environmental reasons. Plowing to break up dense grass

mats has also been shown to be beneficial for seedling performance elsewhere (Otsamo

et al. 1995; du Toit et al. 2010). However, the same method may not show similar effects

during initiation of the second rotation (Smith et al. 2001).

Forest restoration of dry sites with focus on Mediterranean ecosystems

Mediterranean ecosystems in Europe have been modified and degraded by human activities

such as timber cutting, agricultural land use and extensive grazing in combination with

834 New Forests (2012) 43:825–848

123

natural factors such as soil erosion and fire for thousands of years. This has led to severe

land degradation over large areas (Albaladejo 1990; Blondel and Aronson 1999). In

addition, in similar ecosystems in California, soil compaction is commonplace in areas

subjected to continuous grazing for approximately 150 years (McCreary and Canellas

2005). Water is a limited resource and these environments are characterized by periods of

severe moisture stress and high radiation. Soils are also poorly developed, shallow

(20–40 cm), stony, poor in organic matter and sometimes with impermeable horizons

(Albaladejo 1990; Puigdefabregas 1998). The restoration goals in these areas are to protect

soils from erosion, increase soil fertility and to restore natural vegetation and forests.

During the early phase of forest restoration on these sites it is important to establish

regeneration with uniform seedling performance to facilitate management and avoid

problems such as patches of soil erosion. However, the semiarid conditions may cause low

seedling survival due to drought, resulting in non-uniform stocking and low restoration

success. In Mediterranean areas, the purpose of MSP during restoration is therefore to

improve deep penetration of roots, increase soil infiltration of water and nutrient avail-

ability, reduce water runoff and to control competing vegetation (Roldan et al. 1996;

Querejeta et al. 2001; Barbera et al. 2005; Palacios et al. 2009). Traditional MSP methods

include subsoiling (i.e. deep ripping to 50–70 cm soil depth) (Fig. 5), mechanical terracing

(2–3 m wide in combination with subsoiling) and mechanical and manual holes for

planting (ca 40 9 40 9 40 cm) (Rojo et al. 2002). The most intense MSP method is

terracing, which was commonly used from 1970 until the end of the 1980s, but it is

controversial from an environmental point of view and is rarely used nowadays (see

environmental impacts).

Subsoiling MSP or mechanical preparation of planting holes (spot tillage) has been

shown to improve performance of planted seedlings due to increased rooting depth that

enables roots to reach sustainable water reserves (Querejeta et al. 2000, 2001; Castillo et al.

2002). The combination of using high quality seedlings following nutrient loading,

planting early in the season and using subsoiling seems to be an efficient strategy for

improved seedling performance (Palacios et al. 2009) (Fig. 6). Through vigorous root

growth, seedlings may reach sustainable water reserves before any drought later in the

Fig. 5 A Quercus ilex plantation following subsoiling (continuous MSP) on former agricultural field insouthern Spain. Photo by Rafael M. Navarro

New Forests (2012) 43:825–848 835

123

season. However, MSP methods in Mediterranean countries are still a controversial issue

between foresters and ecologist (see environmental impacts). At sites already colonized

with herbaceous vegetation, disk ridging and disk harrowing (similar to disk trenching),

may be negative for seedling growth due to stimulated re-colonization of that vegetation

(Varelides and Kritikos 1995). The future use of alternative and less intense MSP methods

such as intermittent subsoiling and small terracing and use of smaller machines will

hopefully minimize the undesirable and irreversible impact that conventional soil prepa-

ration techniques may produce (Navarro-Garnica 1975; Maestre and Cortina 2004; Cortina

et al. 2011).

Similar land degradation as in Mediterranean ecosystems also has occurred elsewhere,

e.g. in Asia including China and in Africa (Stanturf and Madsen 2005; Perrow and Davy

2008). Vegetation in these arid areas is fragile and vulnerable to degradation. Increasing

population along with the high demand for more food for humans and animals and fuel

wood in combination with drought and soil erosion has sometimes led to desertification.

Reclamation of mine sites

Restoration of sites where surface mining of e.g. bauxite, coal and lignite, have been

completed is increasingly common throughout the world (Parotta and Knowles 2001;

Sharma et al. 2004; Kew et al. 2007; Zipper et al. 2011). Exposed spoils and waste dumps

may contain toxic elements and these sites are prone to wind and water erosion, and water

runoff that may reduce water quality in adjacent water bodies (Bradshaw and Huttl 2001).

Restoration is thus important, and forests on many of these sites have largely been replaced

by persistent herbaceous species. However, during the last decades, forest restoration has

often been preferred since forests provide more ecosystem services such as timber, carbon

Fig. 6 Mean survival rate (topbox) and height growth (lowerbox) after 2 years of highfertilized and low fertilizedQuercus ilex seedlings planted inearly season, middle season andlate season with two types of soilpreparation—manually preparedholes (H) and subsoiling (S).Redrawn from Palacios et al.(2009) where the experiment wasconducted in southern Spain

836 New Forests (2012) 43:825–848

123

storage, watershed and water quality protection, and habitats for pre-disturbance plants and

fauna (Zipper et al. 2011). Forest restoration of former mine sites is a difficult task. In

many places the topsoil is removed and buried under masses of spoil and waste dump; soil

chemistry and texture may not be suitable for sustainable tree growth and the operations

and handling of the soils many times results in soil compaction (Showalter et al. 2010).

When mine soils become compacted, mine operators are advised to loosen them using

subsoiling equipment prior to plating trees (Zipper et al. 2011). There are many examples

of improved performance of planted seedlings following subsoiling on such sites. For

example, Skousen et al. (2009) showed that subsoiling significantly improved survival and

growth of Juglans nigra, Liriodendron tulipifera, Quercus rubra and Prunus serotina7 years following planting. However, in the same study there was no improved plant

performance of Fraxinus americana following the same treatment. Others have also shown

better plant development after subsoiling, and sometimes after extended time-periods

(Ashby 1996). Similar to restoration in Mediterranean ecosystems, higher survival and

growth after subsoiling is probably due to better conditions for root growth and more

available water caused by improved infiltration and collection of water. Due to dense and

hard sub-soil horizons, deep subsoiling (1.5 m) may sometimes be necessary for improved

plant performance (Kew et al. 2007). Kost et al. (1998) found that subsoiling to 30 cm soil

depth was not enough to loosen subsoil horizons on coal mine soil in eastern USA.

In some cases, research has questioned the long-term efficiency of subsoiling because

re-compaction of soils at mine sites may occur (Moffat and Bending 2000). Alternative

methods that may better avoid soil compaction include loose tipping of soils (similar to

mounding) (Fig. 7) or complete cultivation of soils using an excavator. Moffat and

Fig. 7 Planting operation following loose dumping of soil during mine reclamation in eastern USA. Photoby Justin Schmal

New Forests (2012) 43:825–848 837

123

Bending (2000) showed results from several sites in the UK where loose tipping in mounds

and complete cultivation resulted in improved seedling performance 5 years following

planting compared to subsoiling. In the same study, root development in terms of increased

rooting depth and occupied soil volume was also improved. As in many cases with MSP

methods, they also found differences in responses to treatments between tree species.

Stand rehabilitation

The goal of stand rehabilitation through natural regeneration, artificial planting or seeding

with native tree species is often to restore forest structure in e.g. homogenous forest

monocultures (Harrington 1999). Increased biodiversity, stand stability, adaptation to

climate change and promotion of a sustainable economy are reasons for such restoration

(e.g. Hartley 2002; Spiecker et al. 2004).

Following disturbances such as major blowdowns, seeds of various tree species nor-

mally germinate in the resulting pits and mounds created by uprooted trees (e.g. Ilisson

et al. 2007). Similar conditions can be created with different MSP methods to improve

natural regeneration following clear-cutting or beneath shelterwoods. Scarification and

mounding may improve natural regeneration of wind-dispersed tree species such as Betulaspp., especially during years with good seed production (Prevost 1997; Karlsson 2001;

Karlsson and Nilsson 2005; Elie et al. 2009). In contrast, for tree species that depend on

well-developed advanced regeneration, MSP may damage regeneration by reducing

seedling density (e.g. Prevost 1992).

In Europe, the conversion of pure conifer stands into mixed stands with broadleaves has

been a major challenge during the last decades (Spiecker et al. 2004) (Fig. 2). Scarification

on clear-cuts somewhat improved survival of planted Betula seedlings, but had little effect

on growth (Bergquist et al. 2009). In the same study, survival of Quercus seedlings was not

improved by MSP. On the other hand, mounding may improve both survival and growth,

and in some cases this MSP method has been equally effective as repeated herbicide

treatment to increase growth in seedlings (Lof et al. 2006) (Fig. 8). Similar findings have

been reported during restoration of Pinus palustris stands in North America (Knapp et al.

Fig. 8 Mean Quercus roburseedling dry weight 3 yearsfollowing planting in fourtreatments. The treatments were:control (C), repeated herbicidetreatment of ground vegetation(H), mounding site preparation(MSP) and the combination ofmounding site preparation andrepeated herbicide treatment(MSP?H). Redrawn from Lofet al. (2006) where theexperiment was conducted insouthern Sweden

838 New Forests (2012) 43:825–848

123

2006). Mounding has also successfully been used during planting for restoration of coastal

swamp forest in South America (Zamith and Scarano 2010). Underplanting with valuable

tree species under a shelterwood or partial overstory has been proposed as a method that

meets more objectives during rehabilitation than planting on clear-cuts (Paquette et al.

2006). Here light is the most limiting factor although root competition from shelterwood

trees is an additional factor that influences seedling performance (Petritan et al. 2011),

especially on xeric sites or in Mediterranean climates. In general, responses in seedling

survival and growth are therefore less pronounced following MSP beneath shelterwoods

compared to clear-cuts (e.g. Gemmel et al. 1996; Lof 2000).

Direct seeding of large seeded and valuable tree species for stand rehabilitation has been

proposed as a means to lower regeneration costs (e.g. Gonzalez-Rodrıgues et al. 2011).

Similar to planting, mounding may improve for example growth in Quercus seedlings

following direct seeding (Lof and Birkedal 2009). Granivorous rodents, however, may find

and consume buried seeds and their predation may put the whole regeneration operation at

risk (Dey et al. 2008). Scarification of normal intensity does not protect the seeds from this

predation (Birkedal et al. 2009). MSP through intermittent mounding or intensive soil

scarification (similar to blading) may reduce the predation somewhat (Birkedal et al. 2010).

However, even more intensive MSP methods should probably be used to protect regen-

eration from predation (Johnson 1981), which raises concerns from an environmental point

of view.

Economic efficiency

The management goals for forest restoration differ from those for reforestation. In the latter

case, stand replacement is most often scheduled to minimize the period between harvest

and crop tree establishment. Here, MSP is important because it is designed to improve tree

establishment and early survival and growth and many times results in larger land

expectation values although economics are affected by rotation length and discount rate

(Hawkins et al. 2006). During forest restoration early survival and growth is still important,

but are normally not optimized in order to simultaneously achieve other goals such as

improved soil protection and biodiversity; therefore economic evaluations are more

complex. In addition, subsidies many times play an important role during restoration

projects, which also increases complexity in economic evaluations and influences the

choice of methods during implementation. However, ecological and economic goals are

not mutually exclusive during forest restoration (Stanturf et al. 2001). Public landowners

and small non-industrial landowners are not only interested in multi-purpose management

including biodiversity and soil protection, but are also interested in timber production. The

latter may be capable of producing a financial return sufficient to ensure the long term

maintenance of the restored sites (Bradshaw and Huttl 2001). Therefore, efficient estab-

lishment of tree stands and woody vegetation during forest restoration is important as

failure leads to unnecessary costs and potentially to restoration failures or extended time

periods for achieving restoration success.

The body of research on MSP during reforestation for increased production is large and

diverse and the majority of this literature focuses on operational, biological and ecological

aspects of MSP. In contrast, research examining the economics of MSP is far less

developed (Hawkins et al. 2006; Ahtikoski et al. 2010), and there is even less research on

economic efficiency following MSP during forest restoration. However, some knowledge

from reforestation situations can also be used during forest restoration. For example, Uotila

New Forests (2012) 43:825–848 839

123

et al. (2010) compared the economic results from two different MSP methods in Finland.

They used several interest rates and found that an intermittent method (spot mounding)

resulted in lower total management costs compared to disc trenching. Although disc

trenching is a less expensive method compared to spot mounding, the latter method

resulted in less costs for young stand operations to control unwanted vegetation. Fur-

thermore, spot mounding resulted in higher incomes from first commercial thinnings due to

better growth in seedlings and trees. In another study conducted in shelterwoods of dif-

ferent densities by Granhus and Fjeld (2008), the time consumption for manual planting

between two intermittent methods, patch scarification and inverting MSP (or hole plant-

ing), and an untreated control were compared. They found that MSP greatly reduced the

time required for planting, i.e. greater ease of planting, and that the largest reduction

occurred with the inverting MSP. A similar conclusion was drawn by Suadicani (2002)

during conversion of Norway spruce stands although there were no differences between

various MSP methods.

Other goals, e.g. biodiversity or soil protection, may be of higher or equal priority to

stand productivity of timber during forest restoration. Few studies involving MSP have at

the same time taken into account effects on seedling performance, economical costs and

the ecological consequences. However, in a multi-criteria evaluation during restoration of

Pinus nigra forests in Spain, Espelta et al. (2003) concluded that subsoiling (ripping) alone

without other soil or vegetation management methods was the top ranked alterative as it

combined low economic cost with low ecological impact but still resulted in reasonably

good survival and growth of seedlings. In a study by Gan et al. (1998), respondents

indicated greater willingness-to-pay for stands created without site preparation, i.e. less

homogenous stands. Although herbicides, and not MSP for soil treatment, were included in

the latter study, the result indicates that the non-timber values might be higher in stands

without intensive vegetation management control.

A preferable method to establish trees during forest restoration may be to use MSP in a

few spots in combination with the use of large seedlings. Even if larger seedlings cost more

to produce and plant, large seedlings normally have a significantly higher survival fol-

lowing planting and compete better with vegetation than smaller seedlings (e.g. Jobidon

et al. 2003; Cuesta et al. 2010; Dey et al. 2008). Probably, less intensive MSP could be

used in combination with larger seedlings as well. However, to our knowledge little

research and no economic evaluations have been conducted using such alternative man-

agement methods during forest restoration.

Impacts on environment

Depending on the intensity, site preparation using MSP can lead to impact over relatively

large areas and deep soil disturbance. In general, there is a loss of carbon from soil

following MSP, which tends to increase with increased soil disturbance (Jandl et al. 2007

and references therein). At treated areas, organic matter is often mixed with or buried

beneath the mineral soil and, thus, exposed to different conditions of decomposition and

mineralization compared to untreated areas. The soil disturbance changes the microclimate

and stimulates the decomposition of soil organic matter (Johansson 1994). However,

performance of seedlings and carbon sequestration are also favored by MSP in most

studies, and this may balance or outweigh the carbon losses from soils in the overall

ecosystem response (Jandl et al. 2007). Therefore, to avoid unnecessary losses of carbon

from soils following MSP, the chosen technique and intensity is important. Hence, to

840 New Forests (2012) 43:825–848

123

minimize soil disturbance, intermittent or directed MSP should be chosen instead of

continuous methods.

There is a risk that at later stages of stand development, growth may be retarded because

of leaching of elements from the ecosystem due to MSP (Lundmark 1988). Although

nutrient pools may be affected (e.g. Yildiz et al. 2009), there are few studies that have

shown that MSP has negatively affected long-term productivity of crop trees. However, at

sites where there is a risk for soil erosion, sites with low fertility and with low soil organic

matter, harvesting and MSP in combination with wet weather may cause major physical

disturbances, soil compaction and erosion (Miwa et al. 2004). For example in steep forest

lands in northern Spain, mechanized forest operations including down-slope plowing

significantly increased soil losses, with effects on site productivity (Edeso et al. 1999).

Terracing has also been reported to cause soil erosion in Mediterranean areas (Maestre and

Cortina 2004). Most MSP methods can cause soil erosion if not carefully implemented and

adapted to the specific site characteristics and climate (Alcazar et al. 2002). In addition,

without protection zones of trees and vegetation and careful harvesting and scarification,

leached nutrients and organic matter may also affect water quality in adjacent streams

(Ahtiainen 1992).

Mechanical site preparation may influence biodiversity, at least in the short-term, but

effects on biodiversity following MSP are probably more dependent on management of e.g.

the overstory or land use history in the surrounding areas than on MSP itself. In Norway

spruce dominated landscapes in southern Sweden, exposed mineral soil after disc trenching

increased the amount of regenerated broadleaved tree species during the first years after

treatment (Karlsson and Nilsson 2005). However, management such as pre-commercial

thinning and any browsing pressure will probably influence the dynamics of tree species

more through the rotation. Simmons et al. (2011) reported that mounding in temporarily

flooded environments created several microhabitats that created a more diverse plant

community. Haeussler et al. (2004) stressed, however, that site preparation may alter plant

diversity initially but over time the plant communities showed a strong resilience against

disturbance from MSP. The effect of MSP on biodiversity is also dependent on the

intensity of site preparation. For example, Espelta et al. (2003) showed that intensive

vegetation control affected the populations and diversity of small mammals in Spain.

Amphibians, in particular, may suffer from intensive site preparation (Hartley 2002). Little

is known about effects of MSP on below-ground fauna. However, Thornton and Matlack

(2002) indicated that it takes a relatively long time (*50 years) for nematode diversity to

recover after intensive soil disturbance.

Forestry practices, especially MSP, have a great potential to damage ancient remains

(Dolk Frojd and Norman 2007). For example, graves, pits in the ground used as hunting

traps, mounds of stones from former agricultural land use, remains from charcoal stacks

and tar piles may all be hidden under the organic soil layer in the forests. Ancient remains

may also be hidden under agricultural topsoils and be damaged by intensive MSP methods

such as deep plowing (Madsen et al. 2005). In forests, Dolk Frojd and Norman (2007)

reported that as much as 43 % of ancient remains were damaged during harvesting

operations in Sweden and that as much as 65 % were damaged when MSP also was carried

out. They also reported that the most serious damages on ancient remains were caused by

MSP. Directed MSP was the least threatening method because the disturbed area was

relatively small and it was easier to avoid ancient remains compared to use of intermittent

or continuous MSP (Torstensdotter Ahlin 2001).

New Forests (2012) 43:825–848 841

123

Conclusions

Forest restoration projects have become increasingly common around the world and

planting trees is almost always a key component. Low seedling performance caused by e.g.

competing vegetation, excess water or drought, or compacted soils may result in restora-

tion failures and various MSP methods using heavy machinery are important tools used to

help counteract this. MSP methods for treatment of soils such as scarification, mounding

and subsoiling are old methods that were introduced in large scale in forestry with the

ready availability of gas and diesel machinery in the 1940s and were most commonly used

during the 1970s and 1980s. The different types of MSP may be implemented in isolation

or in combinations, and with or without herbicides. Due to environmental problems fol-

lowing operations caused by large areas of soil disturbance, MSP methods did get a

somewhat bad reputation. However, recently they have attracted new attention. Mainly

because public acceptance of chemical herbicides is nowadays low and certification sys-

tems generally discourage them, and therefore managers nowadays are looking for alter-

native site preparation practices.

Although MSP methods are regarded as rather weak tools for vegetation control, they

influence both the amount of competing vegetation and the soil conditions. Methods such

as scarification, mounding and subsoiling also lead to multiple interactions on soil physical

and chemical properties that affect plant performance, and it may thus be difficult to

determine the actual cause–effect relationship of any positive seedling responses. Fur-

thermore, these responses may also vary with climatic conditions and site types. The

choice of method in various forest restoration situations may therefore be problematic and

depend on existing abiotic and biotic conditions and expected problems following any

treatment sequence. Finally, most research on MSP and plant performance has been

conducted using a few conifer tree species. However, seedling responses between species

differ and most often other tree species are used during restoration compared to production

forestry. Therefore, the scientific basis for implementing MSP during forest restoration is

rather weak and we need more research for improved understanding. This research should

include a wide spectrum of sites and tree species.

Several management objectives such as timber production, soil protection and increased

biodiversity are normally relevant simultaneously during forest restoration, and therefore

economic evaluations of MSP are complex. However, the few existing reports on the

subject indicate that intermittent or directed MSP methods may be more beneficial from an

economic point of view compared to continuous methods, which are more intensive and

cause larger soil disturbance. Biodiversity may at least temporarily increase following

MSP and there are surprisingly few reports that show a negative impacts following MSP on

long-term productivity of trees. However, especially continuous methods may have large

negative impacts on the environment causing soil erosion, impaired water quality in

adjacent streams and damaged ancient remains. Therefore, MSP methods should be

implemented carefully and adapted to site conditions. With the exception of MSP causing

soil erosion, seedling performance seems to increase with increased soil disturbance.

During any development of new MSP techniques for the future, intermittent or directed

methods to achieve relatively large soil disturbance on a limited area seems to be a

promising starting point. MSP may then lead to a combination of little environmental

impact with improved seedling performance.

Acknowledgments The study was supported by scholarships from Stina Werners fund and from TheSwedish Foundation for International Cooperation in Research and Higher Education offered to M. Lof

842 New Forests (2012) 43:825–848

123

during his visit at Purdue University, USA in the autumn of 2011. Support for R Navarro was received fromthe Spanish Ministry of Science and Innovation, program CGL2008-04503-CO3-02.

References

Ahtiainen M (1992) The effects of forest clear-cutting and scarification on the water-quality of small brooks.Hydobiologia 243:465–473

Ahtikoski A, Virpi A, Kari M (2010) Scots pine stand establishment with special emphasis on uncertaintyand cost-effectiveness, the case of northern Finland. New For 40:69–84

Albaladejo J (1990) Impact of the degradation processes on soil quality in arid Mediterranean environments.In: RubioJL, Rickson J (eds) Strategies to combat desertification in Mediterranean Europe. Com-mission of the European Communities, pp 193–215

Alcazar J, Rothwell RL, Woodward PM (2002) Soil disturbance and the potential of erosion aftermechanical site preparation. North J Appl For 19:5–13

Allen JA, Keeland BD, Stanturf JA, Clewell AF, Kennedy HE Jr (2004) A guide to bottomland hardwoodreforestation. USDA For Serv Southern Res Sta Gen Tech Rep 40

Ammer C, Balandier P, Scott Bentsen N, Coll L, Lof M (2011) Forest vegetation management under debate:an introduction. Eur J For Res 130:1–5

Ashby WC (1996) Red oak and black walnut growth increased with minesoil ripping. Int J Mining ReclamEnviron 10:113–116

Barbera G, Martınez-Fernandez F, Alvarez-Rogel J, Albaladejo J, Castillo V (2005) Short- and interme-diate-term effects of site and plant preparation techniques on reforestation of a semiarid ecosystemwith Pinus halepensis Mill. New For 29:177–198

Barring U (1965) Om flackupptagningens betydelse och nagra andra problem vid plantering av tall och gran.Stud For Suec 24:80

Barring U (1967) Studier av metoder for plantering av gran och tall pa akermark i sodra och mellerstaSverige. Stud For Suec 50:332

Bergquist J, Lof M, Orlander G (2009) Effects of roe deer browsing and site preparation on performance ofplanted broadleaved and conifer seedlings when using temporary fences. Scand J For Res 24:308–317

Berry CR (1986) Subsoiling improves growth of trees on a variety of sites. In: Phillips DR (ed) Proceedingsfourth biennial southern silvicultural research conference. Gen. Tech. Rep. SE-42, USDA ForestService, Southeast Forest Experimental Forest, Asheville, North Carolina. C, USA, pp 360–367

Birkedal M, Fisher A, Karlsson M, Lof M, Madsen P (2009) Rodent impact on establishment of directseeded beech and oak on forest land. Scand J For Res 24:298–307

Birkedal M, Lof M, Olsson GE, Bergsten U (2010) Effects of granivorous rodents on direct seeding of oakand beech in relation to site preparation and sowing date. For Ecol Manage 259:2382–2389

Blondel J, Aronson J (1999) Biology and wildlife of the mediterranean region. Oxford University Press,Oxford

Boateng JO, Heineman JL, Bedford L, Linnell Nemec AF, McClarnon J (2011) Twenty year site preparationeffects on sub-boreal lodgepole pine performance. New For. doi:10.1007/s11056-011-9292-6

Bocio I, Bruno Navarro F, Ripoll MA, Jimenez MN, De Sımon E (2004) Holm oak (Quercus rotundifoliaLam.) and Aleppo pine (Pinus halepensis Mill.) response to different soil preparation techniquesapplied to forestation in abandoned farmland. Ann Sci For 61:171–178

Bradshaw AD, Huttl RF (2001) Future mine site restoration involves a broader approach. Ecol Eng17:87–90

Castillo V, Gonzalez G, Mosch W, Navarro-Cano J, Conesa C, Lopez F (2002) Seguimiento y Evaluacion delos trabajos de restauracion hidrologico-forestal. In: Lopez F (ed) Seguimiento y evaluacion de losefectos sobre el medio natural de la sequıa y los procesos erosivos en la Region de Murcia. Consejerıade Agricultura Agua y Medio Ambiente de la Region de Murcia, Murcia (Spain), pp 166–233

Cogliastro A, Gagnon D, Bouchard A (1997) Is site preparation necessary for bur oak receiving post-planting weed control? Ann Sci For 54:107–116

Cortina J, Amat B, Castillo V, Fuentes D, Maestre F, Padilla F, Rojo L (2011) The restoration of vegetationcover in the semi-arid Iberian southeast. J Arid Environ 12:1377–1384

Cuesta B, Villar-Salvador P, Puertolas J, Jacobs DF, Rey Benayas JM (2010) Why do large, nitrogen richseedlings better resist stressful transplanting conditions? A physiological analysis in two functionallycontrasting Mediterranean forest species. For Ecol Manage 260:71–78

Dey DC, Jacobs D, NcNabb K, Miller G, Baldwin V, Foster G (2008) Artificial regeneration of major oak(Quercus) species in the eastern United States—a review of the literature. For Sci 54(1):77–106

New Forests (2012) 43:825–848 843

123

Dolk Frojd C, Norman P (2007) Uppfoljning av skador pa fornlamningar i skogsmark. Rapport 9. Skogs-styrelsens forlag, Jonkoping

du Toit B, Smith CW, Little KM, Boreham G, Pallet RN (2010) Intensive, site-specific silviculture:manipulating resource availability at establishment for improved stand productivity: a review of SouthAfrican research. For Ecol Manage 259:1836–1845

Edeso JM, Merino A, Gonzalez MJ, Marauri P (1999) Soil erosion under different harvesting managementin steep forestlands from northern Spain. Land Degrad Dev 10:79–88

Elie JG, Ruel JC, Lussier JM (2009) Effect of browsing, seedbed, and competition on the development ofYellow birch seedlings in high graded stands. North J Appl For 26:99–105

Espelta JM, Retana J, Habrouk A (2003) An economic and ecological multi-criteria evaluation of refor-estation methods to recover burned Pinus nigra forests in NE Spain. For Ecol Manage 180:185–198

Gan J, Kolison SH Jr, Miller JH, Hargrove TM (1998) Effects of site preparation on timber and non-timbervalues of loblolly pine plantations. For Ecol Manage 107:47–53

Gardiner ES, Russell DR, Oliver M, Dorris LC Jr (2002) Bottomland hardwood afforestation: state of theart. In: Holland MM, Warren ML Jr, Stanturf JA (eds) Proceedings of a conference on sustainability ofwetlands and water resources: how well can riverine wetlands continue to support society into the 21stcentury? Gen Tech Rep SRS-50. USDA Forest Service, Southern Research Station, Asheville,pp 75–86

Gemmel P, Nilsson U, Welander T (1996) Development of oak and beech seedlings planted under varyingShelterwood densities and with different site preparation methods in southern Sweden. New For12:141–161

Gonzalez-Rodrıgues V, Navarro-Cerillo RM, Villar R (2011) Artificial regeneration with Quercus ilex L.and Quercus suber L. by direct seeding and planting in southern Spain. Ann Sci For 68:637–646

Granhus A, Fjeld D (2008) Time consumption of planting after partial harvest. Silva Fennica 42:49–61Groninger JW, Baer SG, Babassana D-A, Allen DH (2004) Planted green ash (Fraxinus pennsylvanica

Marsh.) and herbaceous vegetation reponses to initial competition control during the first 3 years ofafforestation. For Ecol Manage 189:161–170

Gwaze D, Johanson M, Hauser C (2007) Long-term soil and shortleaf pine responses to site preparationripping. New For 34:143–152

Haeussler S, Bartemucci P, Bedford L (2004) Succession and resilience in boreal mixedwood plant com-munities 15-16 years after silvicultural site preparation. For Ecol Manage 199:349–370

Harrington CA (1999) Forests planted for ecosystem restoration or conservation. New For 17:175–190Hartley MJ (2002) Rationale and methods for conserving biodiversity in plantation forests. Foe Ecol

Manage 155:81–95Hawkins CBD, Steele TW, Letchford T (2006) The economics of site preparation and the impacts of current

forest policy: evidence from central British Columbia. Can J For Res 35:482–494Horsley SB (1985) Reforestation of orchard stands and savannahs on Pennsylvania’s Allegheny Plateau.

North J Appl For 2:22–26Horsley SB (1991) Using Roundup and Oust to control interfering understories in Allegheny hardwood

stands. USDA For Serv Northeastern For Exp Sta Gen Tech Rep 148:281–290Ilisson T, Koster K, Vodde F, Jogiste K (2007) Regeneration development 4-5 years after a storm in Norway

spruce dominated forests, Estonia. For Ecol Manage 250:17–24Iverson LR, Hutchinson TF, Prasad AM, Peters MP (2008) Thinning, fire, and oak regeneration across a

heterogeneous landscape in the eastern U.S.: 7-year results. For Ecol Manage 255:3035–3050Jacobs DF, Ross-Davis AL, Davis AS (2004) Establishment success of conservation tree plantations in

relation to silvicultural practices in Indiana, USA. New For 28:23–36Jandl R, Lindner M, Vesterdal L, Bauwens B, Baritz R, Hagedorn F, Johnson DW, Minkkinen K, Byrne KA

(2007) How strongly can forest management influence soil carbon sequestration? Geoderma137:253–268

Jobidon R, Roy V, Cyr G (2003) Net effect of competing vegetation on selected environmental conditionsand performance of four seedling stock sizes after eight years in Quebec (Canada). Ann Sci For60:691–699

Johansson M-B (1994) The influence of soil scarification on the turn-over rate of slash needles and nutrientrelease. Scand J For Res 9:170–179

Johnson RL (1981) Oak seeding—it can work. South J Appl For 5:28–33Kabrick JM, Dey DC, Van Sambeek JW, Wallendorf M, Gold MA (2005) Soil properties and growth of

swamp white oak and pin oak on bedded soils in the lower Missouri River floodplain. For Ecol Manage204:315–327

Karlsson A (1996) Initial seedling emergence of hairy birch and silver birch on abandoned fields followingdifferent site preparation regimes. New For 11:93–123

844 New Forests (2012) 43:825–848

123

Karlsson M (2001) Natural regeneration of broadleaved tree species in southern Sweden—effects of sil-vicultural treatments and seed dispersal from surrounding stands. Silvestria 196, PhD Thesis, SLU,Alnarp.

Karlsson M, Nilsson U (2005) The effects of scarification and shelterwood treatments on naturally regen-erated seedlings in southern Sweden. For Ecol Manage 205:183–197

Kew GA, Mengler FC, Gilkes RJ (2007) Regolith strength, water retention, and implications for ripping andplant root growth in bauxite mine restoration. Restor Ecol 15:S54–S64

Knapp BO, Wang GG, Walker JL, Cohen S (2006) Effects of site preparation treatments on early growthand survival of planted longleaf pine (Pinus palustris Mill.) seedlings in North Carolina. For EcolManage 226:122–128

Kost DA, Vimmerstedt JP, Brown JH (1998) Topsoiling, ripping, and fertilizing effects on tree growth andnutrition on calcareous minesoils. For Ecol Manage 103:307–319

Kramer PJ, Boyer JS (1995) Water relations of plants and soils. Academic Press, San DiegoLhotka JM, Zaczek JJ (2003) Effects of scarification disturbance on the seedling and midstory layer in a

successional mixed-oak forest. North J Appl For 20:85–91Lof M (2000) Influence of patch scarification and insect herbivory on growth and survival in Fagus sylvatica

L., Picea abies L. Karst. and Quercus robur L. seedlings following a Norway spruce forest. For EcolManage 134:111–123

Lof M, Birkedal M (2009) Direct seeding of Quercus robur L. for reforestation: The influence of mechanicalsite preparation and sowing date on early growth of seedlings. For Ecol Manage 258:704–711

Lof M, Isacsson G, Rydberg D, Welander NT (2004) Herbivory by the pine weevil (Hylobius abietis L.) andshort-snouted weevils (Strophosoma melanogrammum Forst. and Otiorhynchus scaber L.) during theconversion of a wind-thrown Norway spruce forest into a mixed-species plantation. For Ecol Manage190:281–290

Lof M, Rydberg D, Bolte A (2006) Mounding site preparation for forest restoration: survival and growthresponses in Quercus robur L. seedlings. For Ecol Manage 232:19–25

Londo AJ, Mroz GD (2001) Bucket mounding as a mechanical site preparation technique in wetlands. NorthJ Appl For 18(1):7–13

Lundmark JE (1988) Skogsmarkens ekologi: Standortsanpassat skogsbruk. Del 2, tillampning. Skogssty-relsen, Jonkoping

Luoranen J, Rikala R (2012) Field performance of Scots pine (Pinus sylvestris L.) seedlings planted in disctrenched or mounded sites over an extended planting season. New For. doi:10.1007/s11056-012-9307-y

Madsen P (1995) Effects of soil water content, fertilization light, weed competition and seedbed type onnatural regeneration of beech (Fagus sylvatica). For Ecol Manage 72:251–264

Madsen P, Jensen FA, Fodgaard S (2005) Afforestation in Denmark. In: Stanturf JA, Madsen P (eds)Restoration of boreal and temperate forests. CRC Press, Boca Raton, pp 211–224

Maestre F, Cortina J (2004) Are Pinus halepensis plantations useful as a restoration tool in degradedsemiarid Mediterranean areas? For Ecol Manage 198:303–317

Margolis HA, Brand DG (1990) An ecophysiological basis for understanding plantation establishment. CanJ For Res 20:375–390

McCreary D, Canellas I (2005) Restoration of oak woodlands in Mediterranean ecosystems. In: Stanturf JA,Madsen P (eds) Restoration of boreal and temperate forests. CRC Press, Boca Raton, pp 253–266

Miller JH (1993) Oak plantation establishment using mechanical, burning and herbicide treatments. In:Loftis DL, McGee CE (eds) Oak regeneration, serious problems, practical recommendations. Gen TechRep SE-84, Southeastern Forest Experiment Station USDA Forest Service, Asheville, NC, USA,pp 264–289

Miwa M, Aust WM, Burger JA, Patterson SC, Carter EA (2004) Wet-weather harvesting and site prepa-ration effects on coastal plain sites: a review. South J Appl For 28:137–151

Moffat AJ, Bending NAD (2000) Replacement of soil and soil-forming materials by loose tipping inreclamation to woodland. Soil Use Manage 16:75–81

Munson AD, Margolis HA, Brand DG (1993) Intensive silvicultural treatments: impacts on soil fertility andplanted conifer response. Soil Sci Soc Am J 57:246–255

Navarro Cerrillo RM, Frageiro B, Ceaceros C, del Campo A, de Prado R (2005) Establishment of Quercusilex L. subsp. ballota Desf. Samp. using different weed control strategies in Southern Spain. Ecol Eng25:332–342

Navarro-Garnica M (1975) Tecnicas de Forestacion. Ministerio de Agricultura, ICONANeckelmann J (1976) Soil preparation and plant development on sandy soils in Juland. Dansk Skovforenings

Tidskrift 61:4–34

New Forests (2012) 43:825–848 845

123

Nilsson U, Orlander G (1995) Effects of regeneration methods on drought damage to newly planted Norwayspruce seedlings. Can J For Res 25:790–802

Nyland RD (1996) Silvicultural concepts and applications. McGraw-Hill, New YorkOrlander G, Nilsson U (1999) Effects of reforestation methods on pine weevil (Hylobius abietis) damage

and seedling survival. Scand J For Res 14:341–354Orlander G, Gemmel P, Hunt J (1990) Site preparation: a Swedish overview. FRDA report 105. Ministry of

Forests, Research Branch, VictoriaOrlander G, Nordborg F, Gemmel P (2002) Effects of complete deep-soil cultivation on initial forest stand

development. Stud For Suec 213:22Otsamo A, Adjers G, Hadi TS, Kuusipalo J, Tuomela K, Vuokko R (1995) Effect of site preparation and

initial fertilization on the establishment and growth of 4 plantation tree species used in reforestation ofImperata Cylindrica (L) Beauv dominated grasslands. For Ecol Manage 73:271–277

Palacios G, Navarro-Cerrillo RM, del Campo A, Toral M (2009) Site preparation, stock quality and plantingdate effect on early establishment of Holm oak (Quercus ilex L.) seedlings. Ecol Eng 35:38–46

Paquette A, Bouchard A, Cogliastro A (2006) Survival and growth of under-planted trees: a meta-analysisacross four biomes. Ecol Appl 16:1575–1589

Parotta JA, Knowles OH (2001) Restoring tropical forests on lands mined for bauxite: examples from theBrazilian Amazon. Ecol Eng 17:219–239

Patterson WB, Adams JC (2003) Soil, hydroperiod and bedding effects on restoring bottomland hardwoodson flood-prone agricultural lands in North Louisiana, USA. Forestry 76:181–188

Pausas J, Blade C, Valdecantos A, Seva J, Fuentes D, Alloza J, Vilagrosa A, Bautista S, Cortina J, Vallejo V(2004) Pines and oaks in the restoration of Mediterranean landscapes: new perspectives for an oldpractices. Plant Ecol 209:209–220

Perrow MR, Davy AJ (eds) (2008) Handbook of ecological restoration. Volume 2: Restoration in practice.Cambridge University Press, Cambridge

Peters R (1997) Beech forests. Geobotany. Kluwer, Dordrecht, vol 24Petersson M, Orlander G, Nordlander G (2005) Soil features affecting damage to conifer seedlings by the

pine weevil Hylobius abietis. Forestry 78:83–92Petritan IC, von Lupke B, Petritan AM (2011) Effects of root trenching of overstorey Norway spruce (Picea

abies) on growth and biomass of underplanted beech (Fagus sylvatica) and Douglas fir (Pseudotsugamenziesii) saplings. Eur J For Res 130:813–828

Prevost M (1992) Effets du scarifiage sur les proprietes du sol, la croissance des semis et la competition:Revue des connaissances actuelles et perspectives de recherches au Quebec. Ann For Sci 49:277–296

Prevost M (1997) Effects of scarification on seedbed coverage and natural regeneration after a group seed-tree cutting in a black spruce (Picea mariana) stand. For Ecol Manage 94:219–231

Puigdefabregas J (1998) Ecological impacts of global change on drylands and their implications fordesertification. Land Degrad Dev 9:393–406

Querejeta JI, Roldan A, Albaladejo J, Castillo V (2000) Soil physical properties and moisture contentaffected by site preparation in the afforestation of a semiarid rangeland. Soil Sci Soc Am J 64:2087–2096

Querejeta JI, Roldan A, Albaladejo J, Castillo V (2001) Soil water availability improved by site preparationin a Pinus halepensis afforestation under semiarid climate. For Ecol Manage 149:115–128

Rathfon RA, Lichti NI, Swihart RK (2008) Disking and mid- and understory removal following an above-average acorn crop in three mature oak forests in southern Indiana. USDA For Serv Northern Res StaGen Tech Rep P-24, pp 59–69

Rebbeck J (2012) Fire management and invasive plants in oak ecosystems. USDA For Serv Northern ResSta Gen Tech Rep (in press)

Ritari A, Lahde E (1978) Effects of site preparation on physical properties of the soil in a thick-humusspruce stand. Comm Inst For Fenn 92(7)

Rojo L, Garcıa F, Martınez J, Martınez A (2002) Management plan to combat desertification in theGuadalentin River basin: In: Geeson N, Brandt C, Thornes J (eds) Mediterranean desertification: amosaic of processes and responses. Wiley, Chichester, pp 303–317

Roldan A, Querejeta JI, Albaladejo J, Castillo V (1996) Survival and growth of Pinus halepensis Millerseedlings in a semiarid environment after forest soil transfer, terracing and organic amendments. AnnSci For 53:1099–1112

Russell DR Jr, Hodges JD, Ezell AW (1997) An evaluation of hardwood reforestation methods on previ-ously farmed lands in central Alabama. In: Waldrop TA (ed) Proceedings ninth biennial southernsilvicultural research conference. Gen. Tech. Rep. SRS-20. U.S. Department of Agriculture, ForestService, Southern Research Station, pp 272–276

846 New Forests (2012) 43:825–848

123

Seifert JR, Woeste K (2002) Evaluation of four herbicides and tillage for weed control on 1-0 planted treeseedlings. North J Appl For 19(3):101–105

Seifert JR, Pope PE, Fischer BC (1985) The effects of three levels of site preparation on planted swampchestnut oak on a poorly drained site. USDA FS Southern For Exp Sta Gen Tech Rep SO-54, pp 53–56

Self AB, Ezell AW, Londo AJ, Hodges JD (2010) Evaluation of Nuttall oak and cherrybark oak survival byplanting stock and site preparation treatment type in a WRP planting on a retired agricultural site. In:Stanturf JA (ed) Proceedings fourteenth biennial southern silvicultural research conference. Gen TechRep SRS-121. U.S. Department of Agriculture, Forest Service, Southern Research Station, pp 159–163

Sharma KD, Kumar P, Gough LP, Sanfilipo JR (2004) Rehabilitation of a lignite mine-disturbed area in theIndian desert. Land Degrad Develop 15:163–176

Showalter JM, Burger JA, Zipper CE (2010) Hardwood seedling growth on different mine spoil types withand without topsoil amendment. J Environ Qual 39:483–491

Simmons ME, Wu XB, Whisenant SG (2011) Plant and soil responses to created microtopography and soiltreatments in bottomland hardwood forest restoration. Restor Ecol 19:136–146

Skousen J, Gorman J, Pena-Yewtukhiw E, King J, Stewart J, Emerson P, DeLong C (2009) Hardwood treesurvival in heavy ground cover on reclaimed land in West Virginia: mowing and ripping effects.J Environ Qual 38:1400–1409

Smith CW, Little KM, Norris CH (2001) The effect of land preparation at reestablishment on the pro-ductivity of fast growing hardwoods in South Africa. Austr For 64:165–174

Soderstrom V (1976) Analys av markberedningseffekterna vid plantering pa nagra farska hyggen. SverigesSkogsvardsforbunds Tidskrift 2(3):58–333

Soderstrom V, Backe J, Byfalk R, Jonsson C (1978) Jamforelse mellan plantering i jordrabatter och efterandra markberedningsmetoder. Rapporter och uppsatser, nr 11. Institutionen for Skogsskotsel,Skogshogskolan, Umea

Spiecker H, Hansen J, Klimo E, Sterba H, Skovsgaard JP, von Teuffel K (eds) (2004) Norway spruceconversion—options and consequences. EFI Research Report 18. Brill Academic Publishers, Leiden

Stanturf JA (2005) What is forest restoration? In: Stanturf JA, Madsen P (eds) Restoration of boreal andtemperate forests. CRC Press, Boca Raton, pp 3–11

Stanturf JA, Madsen P (eds) (2005) Restoration of boreal and temperate forests. CRC Press, Boca RatonStanturf JA, Schoenholtz SH, Schweitzer CJ, Shepard JP (2001) Ashieving restoration success: myths in

bottomland hardwood forests. Restor Ecol 9:189–200Stanturf JA, Conner WH, Gardiner ES, Schweitzer CJ, Ezell AW (2004) Practice and perspective: recog-

nizing and overcoming difficult site conditions for afforestation of bottomland hardwoods. Ecol Restor22:183–193

Suadicani K (2002) From plantation towards close-to-nature forestry—operational efficiency of shelterwoodregeneration and selection management in Norway spruce. Doctoral thesis. Danish Centre for Forest,Landscape and Planning. Hørsholm, Denmark

Sutton RF (1993) Mounding site preparation: a review of European and North American experience. NewFor 7:151–192

Tamm CO (1991) Nitrogen in terrestrial ecosystems, ecological studies, no 81. Springer, BerlinThiffault N, Roy V (2011) Living without herbicides in Quebec (Canada): historical context, current

strategy, research and challenges in forest vegetation management. Eur J For Res 130:117–133Thiffault N, Titus BD, Moroni MT (2010) Silviculture and planted species interact to influence reforestation

success on a Kalmia-dominated site—a 15 year study. For Chron 86:234–242Thornton CW, Matlack GR (2002) Long-term disturbance effects in the nematode communities of south

Mississippi woodlands. J Nematol 34:88–97Torstensdotter Ahlin I (2001) Forsok med markberedning inom omraden med fossil akermark i vastra

Gotalands, Jonkopings, Kronobergs och Uppsala lan—ett pilotprojekt. Riksantikvariambetet,Stockholm

Uotila K, Rantala J, Saksa T, Harstela P (2010) Effect of soil preparation method on economic result ofNorway spruce regeneration chain. Silva Fenn 44:511–524

Van Lerberghe P, Balleux P (2001) Afforesting agricultural land. Institut Pour le Developpement Forestier,Paris

Varelides C, Kritikos T (1995) Effect of site preparation and fertilization on Pinus pinaster survival andheight growth on 3 sites in northern Greece. For Ecol Manage 77:111–115

Weber N (2005) Afforestation in Europe: lessons learned, challenges remaining. In: Stanturf JA, Madsen P(eds) Restoration of boreal and temperate forests. CRC Press, Boca Raton, pp 121–135

Willoughby I, Balandier P, Scott-Bentsen N, McCarthy N, Claridge J (eds) (2009) Forest vegetationmanagement in Europe: current practice and future requirements. Cost Office, Brussels

New Forests (2012) 43:825–848 847

123

Yildiz O, Esen D, Sarginci M (2009) Long-term productivity effects of different Rhododendron controlmethods in eastern beech (Fagus orientalis Lipsky) ecosystems in western Black sea region of Turkey.Soil Use Manage 25:28–33

Zamith LR, Scarano FR (2010) Restoration of coastal swamp forest in southeast Brazil. Wetlands EcolManage 18:435–448

Zipper CE, Burger JA, Skousen JG, Angel PN, Barton D, Davis V, Franklin JA (2011) Restoring forests andassociated ecosystem services on Appalachian coal surface mines. Environ Manage 47:751–765

848 New Forests (2012) 43:825–848

123